OBJECTIVES: Asthma exacerbations are a primary research endpoint due to the associated cost burden. The generally accepted claims-based algorithm for identifying exacerbations (non-hospitalized) consists of an office visit with an asthma diagnosis and an oral corticosteroid prescription ±7 days of the visit. However, physician-administered biologic therapies for asthma pose specific challenges to this algorithm, due to requiring scheduled visits for administration. We evaluated 2 adjustments to the standard definition to exclude scheduled visits for biologic administration.

METHODS: A retrospective cohort study of commercially insured patients using omalizumab, the biologic with most market exposure, was conducted using medical/pharmacy claims 11/1/14 - 9/30/17. Inclusion criteria were 12-months of continuous enrolment in the year prior to (baseline) and post (follow-up) the 1^st administration, age ≥12, asthma diagnosis during baseline, and ≥2 doses of omalizumab. Patients administered another asthma biologic were excluded. A post-hoc analysis was conducted using 2 adjustments with increasing refinement to the standard exacerbation definition (#1) which does not exclude scheduled visits; (#2) excluding visits using the omalizumab J-code; and (#3) excluding visits within 28 days of omalizumab administration due to the usual 4-week dosing schedule.

RESULTS: 799 patients were identified. Using approach #1, 71.0% of patients in the baseline vs. 63.5% in follow-up had ≥1 asthma exacerbation with a mean(±SD) of 1.7 (1.8) vs. 1.7 (2.1), resulting in no reduction in the exacerbation rate (p=0.837). Using approach #2, 59.6% of patients had ≥1 asthma exacerbation with a mean of 1.5(1.9), resulting in an 11.8% reduction in exacerbation rate (p=0.030). For approach #3, 55.2% of patients had ≥1 asthma exacerbation with a mean of 1.2(1.7), resulting in a 29.4% reduction in exacerbation rate (p<0.001).

CONCLUSIONS: These findings demonstrate that the standard approach for identifying asthma exacerbations for biologic therapies may result in an overestimate of exacerbations due to including scheduled visits for biologic administration.